Patient learning about diabetes through group medical visits in Family Medicine

Purpose/Background: Portland Family Medicine (PFM) initiated group medical visits (GMVs) for diabetic patients in January 2019 to deliver evidence-based diabetes care within a supportive environment, promoting patient learning via shared experiences. The existing literature on GMVs for diabetes care consistently shows improvement in clinical outcomes, when compared to traditional primary care models, but we hoped to examine the effectiveness of our interdisciplinary care model at PFM to see if our learning environment would produce similar benefits. Quality improvement data were obtained to assess the effectiveness of the diabetes groups in increasing participants’ knowledge about diabetes, and in improving clinical outcomes. We hypothesized that participants who improved their diabetes knowledge would have the greatest improvement in clinical outcomes. Methods/Approach: Primary care providers from the Portland and Falmouth Family Medicine offices referred patients with uncontrolled diabetes. Groups were led each week by a physician and a clinical pharmacist. During the initial group, patients completed the Michigan Diabetes Research and Training Center’s (MDRTC’s) Diabetes Knowledge Test. This validated, multiple choice question test, assessed patients’ knowledge about diabetes. Patients’ “pre-test” scores from the initial group were compared to a second administration at least 6 weeks later, to assess patient learning through the group visits. Clinical data collected included patient weights, hemoglobin A1c’s, and other blood work, when clinically appropriate. Groups met once weekly for three visits during the first month, then every other week the second month, then monthly. Paired t-tests were used to compare values at pre- and post-intervention time points. Results: For the 24 patients who participated in diabetes GMVs at Portland Family Medicine between January 2019 and January 2020, average A1c dropped from a baseline of 9.8% to 8.1% at three months (p = 0.005). Of the 24 participants, 17 attended four or more diabetes GMVs (retained cohort), while seven attended fewer than four (lost cohort). The “retained” cohort had a statistically significant drop in mean A1c from 9.7%, to 7.5% at three months (p = 0.004). The “lost” cohort’s mean A1c went from 10.1% to 9.9% at three months (n = 4, too small for p-value calculation). Thirteen patients completed the diabetes knowledge post-test, with an average of 64.2% on the pre-test and 66.9% on the post-test, but this was not a statistically significant increase (p = 0.321). Conclusions: After one year of diabetes group visits at Portland Family Medicine, we have seen the improvement in clinical outcomes seen throughout the literature. Our data also suggest that clinical improvement after 3 months, measured by a statistically significant improvement in average A1c, is not related to increased diabetes knowledge, as measured by the MDRTC’s Diabetes Knowledge Test, suggesting participation alone may be more important than knowledge gained. As the sample continues to grow, further research could explore other clinical or utilization outcomes, including diabetes related ED visits or hospitalizations for group participants, medication adherence, exercise capacity, or progression of disease